---
name: classifying-ai-ambition-with-oar
description: Label every AI initiative Optimize, Amplify, or Reinvent using published evidence, so the language sent to the board matches the work being done.
---

# Classifying AI Ambition With OAR

Label every AI initiative Optimize, Amplify, or Reinvent using published evidence, so the language sent to the board matches the work being done.

## When to use this skill

- A portfolio of AI initiatives is being described as transformation and the results do not support the word.
- A budget defense is coming and each initiative needs an honest level attached.
- The team is trying to pick which single function deserves a Reinvent bet.
- A board deck claims structural change while the underlying work only made an existing step faster.
- An initiative has a level in its name but no stated economic result.

## Inputs to collect

- The full list of live and proposed AI initiatives, including the boring ones. Source: RevOps intake or the AI program tracker.
- For each initiative, the process before and the process after, written as steps. Source: the workflow owner, not the vendor deck.
- For each initiative, whether team composition, workflow ownership, or funnel inputs changed. Source: the org chart and the workflow owner.
- For each initiative, the stated economic result with its metric and its measurement window. Source: finance system or CRM report.
- The language currently used about each initiative in board and investor material. Source: the last board deck.
- Data and process maturity for the owning function. Source: RevOps assessment.

## Process

The audit is one question per initiative: did this change the process, or speed up the existing one (https://www.therevenueaireport.com/frameworks/oar).

1. **Inventory.** List every initiative with its owner and its current spend. An initiative missing from the list cannot be levelled.
2. **Ask the one question.** For each initiative, decide whether the process changed or only got faster. Record the evidence used, not the opinion.
3. **Assign the level.** Apply the three definitions below and write the level next to the initiative.
   - **Optimize.** Same process, made faster. Better sequences, faster call summaries, quicker routing. Real value and the right starting point for a low-maturity team.
   - **Amplify.** Structural change. Team composition moves, workflow ownership moves, and the inputs to pipeline change rather than the throughput. The economics look different at the end of the quarter.
   - **Reinvent.** Rebuild the function from first principles. A process that needed ten people needs two and an AI layer.
4. **Attach the economic proof point.** Write the stated result, the metric, and the window next to each level. An initiative with no economic result gets the label "unproven", not a level.
5. **Compare language to level.** Put the board language and the assigned level side by side and mark every mismatch. The delta between the language and the reality is where reported return disappears (https://www.therevenueaireport.com/frameworks/oar).
6. **Name one Reinvent target.** Choose one function on leverage and exposure. Hand the sequencing of that bet to climbing-oar-levels-with-sling.

## Decision rules

- Level by evidence, not by ambition. The published purpose of OAR is to be honest about which level each initiative is on and to stop reporting Optimize work as transformation (https://www.therevenueaireport.com/frameworks/oar).
- Optimize is not failure. For a team with low data and process maturity, Optimize is the correct starting level. The failure is staying there while claiming otherwise (https://www.therevenueaireport.com/frameworks/oar).
- Treat Reinvent as a bet, not a policy. One function, chosen for leverage and exposure, is the working limit (https://www.therevenueaireport.com/frameworks/oar).
- A level without a stated economic result is a label, not a decision. Mark it unproven and send it back (https://www.therevenueaireport.com/frameworks/oar).
- Use the published realistic gain bands as a sanity check on a claimed level, not as a promise. Optimize is a realistic gain of 10 to 30 percent time savings, Amplify 20 to 40 percent, and Reinvent order-of-magnitude on the right function (https://www.therevenueaireport.com/blog/oar-matrix-ai-value-gap). An initiative claiming Reinvent while reporting a 15 percent time saving is Optimize work with Reinvent language.
- Expect the Reinvent tier to be small. BCG's study of 1,250 companies found only 5 percent of firms generating AI value at scale, outperforming peers at 1.7 times revenue growth, 3.6 times total shareholder return, and 1.6 times EBIT margin (https://www.therevenueaireport.com/blog/oar-matrix-ai-value-gap). A portfolio where most initiatives level as Reinvent has been graded by ambition.
- Do not accept activity gains as a level upgrade. In a single sample of GTM leaders, 87 percent of sales teams reported reps spending more time selling while 13 percent reported higher quota attainment, a 74-point distance (Scale Venture Partners, n=278, https://www.therevenueaireport.com/research/proof-gap). Time freed at the top of the funnel lands on a downstream system that did not get faster.
- Check the level against workflow redesign. McKinsey tested 25 organizational attributes and found fundamentally redesigning workflows had the largest effect on EBIT impact, and at the time only 21 percent had redesigned any workflow (https://www.therevenueaireport.com/research/what-works). An initiative that redesigned no workflow is not Amplify.
- Hand the named Reinvent target to SLING rather than starting it directly. OAR is the target and SLING is the sequencing rule for getting there without breaking production systems (https://www.therevenueaireport.com/frameworks/oar).
- Place the deployment on a level before scoring the result, which is the published order used in every teardown in the Report (https://www.therevenueaireport.com/frameworks/oar).
- The framework publishes no cutoff for how much team composition must move before Optimize becomes Amplify. Set that cutoff with the function owner, write it at the top of the audit, and apply it to every initiative in the same run so the portfolio stays comparable.

## Output requirements

Deliver one row per initiative in this exact shape, because the level and the evidence have to travel together.

| Initiative | Owner | Spend | Process changed or only faster | Level | Economic result, metric, window | Board language today | Mismatch |
|---|---|---|---|---|---|---|---|
| AI call summaries | Enablement dir. | 48k per year | Only faster | Optimize | 3 hours per rep per week, self-reported, 6 months | "AI transformation of coaching" | Yes |
| BDR pod rebuild | VP Sales | 210k per year | Process changed, headcount and ownership moved | Amplify | Pipeline share, quarterly | "efficiency program" | Yes, understated |

Also deliver: the count of initiatives at each level, the list of unproven initiatives, the single named Reinvent target with the reason it was chosen, and the corrected sentence to use with the board.

## Verification loop

Validate the audit before anyone sees a level.

1. Re-read each row and confirm the level was justified by the recorded evidence, not by the initiative's name or its budget.
2. Confirm every row carries either an economic result with a metric and a window, or the label unproven. There is no third state.
3. Confirm no more than one initiative is named as the Reinvent bet, and that its function is named.
4. Check the level distribution against the 5 percent finding (https://www.therevenueaireport.com/blog/oar-matrix-ai-value-gap). If most rows level as Reinvent, the audit graded ambition and must be re-run from step two.
5. Fix the failing rows and repeat checks 1 through 4 on the whole table, because re-levelling one row changes the distribution check.

Only proceed to publish the audit when all four checks pass on the same version of the table. If an initiative's owner disputes a level, record both the assigned level and the disputed level in the row and flag it for human review rather than settling it silently.

## Quality checks

- Every initiative in the tracker appears in the table.
- Every level has evidence written next to it.
- Unproven initiatives are labelled unproven, not levelled optimistically.
- Exactly one Reinvent target is named.
- Every mismatch between board language and assigned level is marked.
- Time savings are never counted as an economic result on their own.
- The level is assigned before any score or verdict is attached to the initiative.
- The claimed level is checked against the published gain bands, not against the budget.
- The audit records the cutoff used for the Optimize to Amplify boundary at the top of the table.
- The corrected board sentence is written out, not described.

## Limitations

- OAR is an aiming tool. It says what level the work is on. It does not say whether the work is a good idea or whether the vendor is right.
- Levels are structured judgment. Two honest reviewers can split on an Amplify or Optimize boundary case, which is why the cutoff has to be written down first.
- The published gain bands come from a blog analysis, not from a controlled study. Use them as sanity checks and say so.
- An audit run once is a snapshot. A level can change when a workflow is redesigned, so re-run it on the same cadence as the budget review.

## Example input

Six live AI initiatives across sales and marketing. Board deck describes the portfolio as "reinventing go-to-market". Reported results are hours saved and content volume. No workflow was redesigned. Illustrative and synthetic, provided to show output shape.

## Example output

Levels: five Optimize, one unproven, zero Amplify, zero Reinvent.

Reason: none of the six changed team composition, workflow ownership, or the inputs to pipeline, so none clears the Amplify definition (https://www.therevenueaireport.com/frameworks/oar). Five report time or volume gains, which are inputs, not outcomes. One reports nothing measurable and is marked unproven.

Mismatch: the board language says reinventing go-to-market and the portfolio levels as Optimize. That is the delta the site identifies as where reported return disappears.

Corrected board sentence: this year the portfolio bought speed in five existing processes at a stated cost, no function has been restructured, and we are naming one function for a Reinvent bet next quarter.

Named Reinvent target: outbound pipeline generation, chosen because it has the highest leverage on the committed pipeline number and the most complete internal data, which is the condition the site's own case evidence turns on.

## Rules of conduct

- Write for a Director, VP, or operator. Short sentences. Explain uncommon terms.
- Separate facts from assumptions. Never hide uncertainty.
- Do not invent numbers, benchmarks, quotes, or customer names.
- Do not send messages, change CRM records, or publish anything unless the user explicitly asks.
- Flag when a decision needs human review.

## Evidence

- https://www.therevenueaireport.com/frameworks/oar
- https://www.therevenueaireport.com/frameworks/sling
- https://www.therevenueaireport.com/frameworks/proof
- https://www.therevenueaireport.com/frameworks/optimization-theater
- https://www.therevenueaireport.com/blog/oar-matrix-ai-value-gap
- https://www.therevenueaireport.com/blog/ai-strategy-sequencing-framework
- https://www.therevenueaireport.com/research/proof-gap
- https://www.therevenueaireport.com/research/what-works

## Cite this framework

Kvarfordt, Jonathan. "OAR." The Revenue AI Report. https://www.therevenueaireport.com/frameworks/oar